Abstract
Economic and social inequalities have generated growing concern and crises across contemporary societies. One of the mechanisms proposed by social sciences to explain the persistence of inequality is the belief in meritocracy, which would legitimize economic disparities based on differences in effort and talent. Despite its wide use as a concept, empirical research on meritocracy is relatively novel and characterized by diverse conceptualization and measures that make the findings and their interpretation rather confusing. Most of the studies in the area have relied upon secondary data to operationalize meritocracy, with a wide variation in the use and interpretation of the same survey items. Taking into account the extant literature that uses measures of meritocracy, this article identifies a series of drawbacks and inconsistencies within and between studies regarding conceptualization and operationalization. Based on this critical analysis, we propose a survey instrument called Perceptions and Preferences for Meritocracy Scale, which we test with confirmatory factor analysis using data from an online survey study (N=2,141). Our results support the proposed conceptual structure, which not only distinguishes between perceptions and preferences, but also between meritocratic and non-meritocratic dimensions. Our discussion highlights the importance of considering these different dimensions in order to advance the study of meritocracy.Economic inequality has become an issue of growing concern around the world. This has been expressed in a series of protests, as well as diverse critical analyses regarding the development of capitalism and its consequences (???; ???). In this context, the study of views, preferences and perceptions of inequality has acquired relevance in the social sciences, in topics such as redistributive preferences (???; ???), the legitimization of economic inequality (???) and the functioning of meritocracy (???; ???; ???; ???).
In general, meritocracy is defined as a system of distribution of resources and rewards based on individual merit, which in its original conception is a combination of talent and effort (???). This traditional conception of merit places in a secondary position the possible interference of structural or non-meritocratic factors, such as inheritance, personal contacts, and luck (???; ???; ???; ???; ???). Social psychology and sociology have studied the characteristics and consequences of beliefs in meritocracy, under the general hypothesis that a greater belief in meritocracy emphasizes the role of individual over structural factors in personal achievements, leading to greater legitimization of inequalities (???; ???; ???; ???). Such research has raised criticism of meritocracy as a moral standard of distribution, arguing that it would be an unfulfilled promise given the still preponderant weight of non-meritocratic elements upon individual status and social mobility(???; ???; ???; ???; ???; ???).
Due to the role that meritocratic beliefs play in the justification of individual achievement or failure in contemporary societies (???; ???), multiple studies have evaluated the relationship between meritocratic beliefs and personal and/or contextual characteristics. For example, some studies have linked meritocracy to the reinforcement of socio-economic, gender, and ethnic stereotypes (???; ???; ???), as well as the effects of meritocratic beliefs in educational (???; ???) and organizational contexts (???; ???).
Most of the studies so far have used indicators from existing standard surveys (as the International Social Survey Programme - ISSP) to measure meritocracy. However, as we will show later, the methods of measuring meritocracy vary extensively among studies. In many cases, similar phenomena are associated with different indicators, and conversely, often different phenomena are measured with similar indicators. Such trends not only limits the comparability of studies, but also the ability to understanding the effects of meritocratic beliefs across different fields.
Based on a critical analysis of different approaches to the measurement of meritocracy to date, this article proposes an instrument to both measure and relate two key aspects in the study of meritocracy: perceptions and preferences. Furthermore, as a second axis of analysis, it considers the elaboration of indicators regarding meritocratic and non-meritocratic aspects, demonstrating that they are not two poles of the same continuum as some previous studies would seem to suggest. The measurement system outlined is oriented to generate an instrument as simple and brief as possible, being suitable to be used in public opinion surveys and allowing to integrate meritocratic beliefs in the study of different social phenomena.
In the following we discuss four critical aspects in studies conceptualizing and measuring meritocracy and that are the basis of our scale proposal.
One recent definition of meritocracy by (???) is the following: “when I discuss meritocracy beliefs, I am referring to citizens’ belief in the importance of hard work relative to structural factors.” (???). In the subsequent operationalization, this is associated with the following question and indicator: “how important you think it is for getting ahead in life: (a) hard work”, scored from 1 to 5 on a likert scale. The assumptions behind such definition are worth discussing in light of the conceptual meaning of meritocracy and its possibilities of operationalization.
The item used by (???) is part of an items’ battery present in several international surveys, usually called “reasons to get ahead”. This battery displays a series of indicators related to what people consider important to get ahead in life: hard work, education, ambition, a wealthy family, the right connections, religion, race, and gender. Therefore, when considering only one of the items, it means that other aspects that could be associated with talent, such as education, would not be deemed as meritocratic. As he points out: “Hard work is arguably the most meritocratic part of Michael Young’s equation: ‘Merit = Intelligence + Effort’, for the simple fact that intelligence itself is influenced by a non-meritocratic factor: who your parents happen to be” (p.5).
In this measure of meritocratic beliefs, we can observe a couple of strong conceptual assumptions: effort would not depend on parental influence, and talent (as innate ability) is not meritocratic (contrary to Michael Youngs’ original conceptualization). This conceptual and measurement-based assumption is found in other studies, that also assume effort to be the main and only aspect of meritocracy (???; ???). Is effort the only dimension behind the concept of meritocracy, or includes other components, such as talent (as in its original conceptualization). The question of whether talent is or is not considered meritocratic is certainly an interesting topic to discuss from a philosophical point of view, but for those working on empirical studies it should be something that face empirical scrutiny. Besides, the consideration of talent as part of meritocracy certainly opens some interesting avenues of research, for instance in studies showing that for the elites meritocracy is more related to talent, whereas effort is something more characteristic of meritocracy of the middle and low classes ([citar Atria ABS]). Actually, in a posterior paper by (???), somewhat contradictory with its previous definition, he considered talent as an indicator of meritocracy.
b. Beliefs?
Several approaches to the empirical study of meritocracy based on public opinion surveys, refer to the concept of beliefs, however different meanings and operationalizations usually underlie this concept. To illustrate this point, lets consider the most commonly used items for operationalizing meritocratic beliefs in survey research (as ISSP), which is the “reasons to get ahead” battery. It consists in a series of items asking “how important you think it is for getting ahead in life” and then lists several factors, such as effort, education, parental wealth, and contacts. Another version of this same battery used in several surveys - sometimes along the previous one - asks about “how important you think it should be …”, then listing the same concepts. Therefore, the question raised here is: Which one of both is a “belief”? What is or what it should be?
The term belief has an ambiguous character in the literature, conceived as “idea-elements” by (???) or “considerations” by (???). As (???) pointed out about the scope of beliefs: “This usage encompasses such more specific social-psychological concepts as values, perceptions, and attitudes” (p.30). Therefore, beliefs cover almost anything related to subjective factors. To this regard, a relevant distinction in the field of inequality beliefs was made by (???): “Perceptions refer to subjective estimates of existing inequality (i.e. thoughts about what is). Beliefs are here defined as normative ideas about just inequality (i.e. thoughts about what should be)”(p.359). Several papers dealing with meritocracy use the term beliefs (i.e. what should be), while actually referring to perceptions (i.e. what is). This occurs not only in (???), but also in (???), in which the term beliefs is used to talk about perceptions, whereas other authors use general terms such as attitudes (???). The first attempt to shed light on this issue was made by (???), who used the question “how important should the number of years spent in education and training be in deciding how much money people ought to earn?” as a proxy for “desired” meritocracy (beliefs). They then determined “perceived” meritocracy, using the questions: “Would you say that in your country, people are rewarded for their efforts?” and “… people are rewarded for their skills?”.
Is the belief in meritocracy a perception or a preference with normative meaning? In order to expand the analytical conceptual framework, we believe that both dimensions should be included in analyses, as proposed by (???). This opens up the possibility of analyzing whether perceptions and preferences are actually related (i.e. have a high correlation), or whether they are independent aspects of the same phenomenon. As (???) have pointed out, “People can believe that outcomes ought to be distributed on the basis of merit and yet vary in their perceptions of whether this is how society currently operates” (p. 435). In other words, normative beliefs should be considered while taking perceptions into account: a strong normative belief in meritocracy may mean something totally different to someone perceiving high meritocracy, when compared to someone perceiving low meritocracy. To avoid the confusion generated by the term “belief”, we propose the terms meritocratic preferences (“what should be”), and meritocratic perceptions (“what is”), as they better reflect the two facets of meritocracy under scrutiny (???).
c. Non-meritocratic aspects
A number of studies have dealt with aspects opposed to meritocracy, as for instance the use of personal advantages (as contacts or having a wealthy family) to get ahead in life. For instance, (???) used an items’ battery listing a number of factors in relation to “How important each should be in deciding pay…” (as (???) for desired meritocracy). They consider factors such as education and responsibility as meritocratic, giving them a value of 1 if considered “essential” in the scale response, whereas factors such as having a family or children are valued 1 when rated as “not important at all” (i.e. reverse coded). The assumption behind this approach is that rejecting a supposed non-meritocratic aspect (as having family and/or children) implies a stronger belief in meritocracy. A similar approach of reverse-coding non-meritocratic items was taken by (???), using the same principles applied in the “Preference for the Merit Principle Scale” (???).
The assumption that meritocratic and non-meritocratic elements are poles of the same continuum was tested by (???) using the “get ahead” perceptions’ battery items mentioned above. They considered education, ambition, and hard work as meritocratic, and other factors such as family wealth and connections as non-meritocratic. Despite making and proving this distinction, however, the authors end up subtracting one dimension from the other, thus coming back to the assumption that they are two poles of the same continuum as (???) did. Similarly, (???) used ISSP indicators for perceived meritocracy and non-meritocracy to build a single score by reverse coding the non-meritocratic items. Therefore, the treatment of non-meritocratic items has been rather inconsistent across studies and the assumption that they are the simple opposite of meritocracy requires further empirical assessment.
d. Accounting for measurement error
Finally, most studies in meritocracy so far have not accounted for the issue of measurement error (???; ???) (Ansolabehere, S., Rodden, J., & Snyder Jr, J. M., 2008), as they use single indicators and/or simple average indexes for measuring meritocracy. Such strategy assumes that the latent construct is measured perfectly (i.e. no error or residual variance) by the indicators chosen, going as far as to propose that “… In choosing this strategy of index construction, we argue that support for meritocracy is not a latent variable” (???). Some advances were made by (???) through conducting a principal component analysis of meritocratic and non-meritocratic dimensions, but somewhat contradictorily, they chose a sum index despite proving a multidimensional latent structure.
Based on the previous assumptions and limitations of the measurements of meritocracy, we propose and test an instrument with the following characteristics:
Multidimensionality, incorporating previous distinctions between preferences and perceptions, as well as between meritocratic and non-meritocratic aspects.
Multiple indicators for each dimension, in order to account for measurement error in a confirmatory factor analysis context.
Based on previous indicators as much as possible, for the sake of keeping comparability between studies.
Brief, as to be used in regular public opinion surveys. In this respect, it differs from the proposal of “Preference for the Merit Principle Scale” (???), as they use 15 items for just one dimension (aside from the problem of reverse-coding non-meritocratic items).
The proposed measurement framework is depicted in Figure 1:
Figure 1: Model of perception and preferences for meritocracy and non-meritocracy
The columns “Perceptions” and “Preferences” represent the distinction between these two concepts, usually confused under the label “beliefs” (???). Perceptions refer to the extent to which people observe that meritocracy works or applies in their society, which relates to items such as “I think hard work is important to get ahead in society”. Preferences refer to normative expectations that are usually linked to a “should” expression (e.g. whether hard work should be related to payment). The rows in Figure 1 consider the distinction between meritocratic and non-meritocratic dimensions (???). These aspects have usually been treated as different ends of the same continuum in previous research, an assumption that requires empirical scrutiny. Non-meritocratic elements usually refer to the use of personal contacts or family advantages to get ahead in life.
Regarding the selection of indicators, most are taken or adapted from previous studies for the sake of comparability. For meritocratic indicators, we use effort and talent as the main components of the traditional concept of merit as defined by (???), whereas for non-meritocratic dimensions, we use wealthy parents and good contacts. Descriptions of the specific items are outlined in the methodology section.
The research hypotheses behind this conceptualization and measurement model are as follows:
\(H_1\). The perception of meritocracy is a latent variable based on indicators of the importance attributed to talent and effort to get ahead in life.
\(H_2\). The perception of non-meritocracy is a latent variable that derives from indicators related to agreement with the statement that people with contacts and rich parents manage to get ahead.
\(H_3\). Meritocratic preferences are a latent variable based on the normative value of effort and talent.
\(H_4\). Non-meritocratic preferences are a latent variable based on the normative value of the use of personal contacts and having wealthy parents.
pacman::p_load(knitr)
The analysis is organized in three studies. Study 1 serves as a background to our instrument proposal by attempting to operationalize the meritocracy dimensions with available secondary data. For this we use data from the International Social Survey Programme (ISSP). Although we are aware that the available indicators in this dataset do not allow a precise operationalization of the conceptual model, we know that this dataset is the most widely used in comparative meritocracy studies (as well as overall subjective inequality) and therefore it is certainly useful to look at its potentials and limitations to this regard. Study 2 is the core of this paper and consists of the proposal of a new scale for measuring meritocratic perceptions and preferences. Finally, in Study 3 we perform additional validity analysis of the meritocracy scale.
Studies that attempt to characterize and compare societies by their support for meritocratic beliefs have used different approaches. As most studies use secondary survey data, they tend to assume that the available indicators represent an underlying meritocratic construct. A review of these studies reveals several non-tested assumptions, as well as the use of similar indicators to represent different constructs and dimensions of meritocracy. As the existence of heterogeneous approaches certainly has consequences for the advancement of the study of meritocracy, this paper presented a comprehensive conceptual framework for the empirical study of meritocracy, building upon previous research. This framework was then tested against new survey data.
We identified four critical aspects regarding the measurement of meritocracy in previous studies: unidimensionality, the ambiguous use of the term “beliefs”, the use of non-meritocratic indicators as opposed to meritocratic indicators, and the consideration of measurement error. The proposed 8-item scale, “Perceptions and Preferences for Meritocracy”, was designed and tested in order to deal with these four issues. Our results indicate that perceptions and preferences seem to be two related but different dimensions, often confounded in previous research under the label of “beliefs”. Meritocratic and non-meritocratic dimensions do not appear to constitute poles of the same continuum, as some previous studies have assumed. Regarding the possible effects of items order in the estimation of latent variable, we tested three different order versions and found evidence that suggest the use of the randomized version of the scale. Furthermore, we found evidence of convergent validity as well as measurement invariance that approach well the requirement of comparability between samples.
The four-dimensional framework and structure of the Perceptions and Preferences for Meritocracy scale opens several avenues for future research. For instance, distinguishing perceptions from preferences will allow us to evaluate the extent to which different societies are accustomed to, or satisfied with, the perceived level of meritocracy, in terms of differences between what is perceived and what is preferred. Additionally, given that non-meritocratic factors are not necessarily related to meritocratic ones, our framework makes it possible to assess the perceived legitimacy of practices such as the use of personal contacts and their interference (or not) with meritocratic ideals in different societies. The consideration of structural factors, both individual and societal, will allow us to advance future hypotheses relating to meritocratic legitimacy to a more comprehensive level than the current one, which is based solely on meritocratic perceptions. Furthermore, the impacts of different configurations of the four-dimensional framework on practices and behaviors such as corruption, civic involvement, and political alignment, is an area that requires additional research. Such future agendas could be specially relevant in times of economic crisis and growing inequalities, that could entail changes in the legitimation of the current distributive structure based on meritocratic ideals.
| Sample | CEP | |
|---|---|---|
| Gender | ||
| Men | 49,82% | 50,52% |
| Women | 50.18% | 49,47% |
| Age | ||
| 18 - 24 | 18,55% | 18,17% |
| 25 - 34 | 18,86% | 17,48% |
| 35 - 44 | 19.09% | 19,98% |
| 45 - 54 | 17,96% | 19,23% |
| 55 - or more | 25,54% | 25.11% |
| Education | ||
| Primary or less | 2,93% | 15,88% |
| High school | 43,23% | 37,04% |
| Non university | 32,63% | 28,93% |
| university or more | 21,21% | 18,13% |